Digigyan SEO Toolkit

Changelog

1.6.5

• Removed: CSS/JS file-based minification feature (wrote .css/.js files to uploads, which is not permitted by the WordPress Plugin Directory guidelines). Browser caching and GZIP .htaccess rules and JS defer are unaffected.

1.6.4

• Fix: JSON-LD </script> escaping is now case-insensitive (preg_replace with the i flag) so mixed-case variants such as </SCRIPT> can no longer break out of the JSON-LD block; the previous str_replace chain was also double-escaping already-escaped sequences.

1.6.3

• Renamed: Plugin is now Digigyan SEO Toolkit (previously Bulk SEO Toolkit) — all internal slugs, upload paths, and text domains updated accordingly.
• Improvement: Admin CSS and JavaScript are now properly enqueued via the admin_enqueue_scripts hook using wp_register_style / wp_add_inline_style and wp_register_script / wp_add_inline_script, replacing the previous inline <style> / <script> output.
• Improvement: Find & Replace, Canonical, and Schema batch scripts now pass their nonce and AJAX URL through wp_localize_script instead of inline JS variables.
• Fix: Removed the upgrader_package_options filter that was no longer needed.
• Fix: JSON-LD output now escapes </script> sequences inside the schema JSON to prevent HTML parser breakout.
• Fix: Admin path resolution now uses WP_CONTENT_DIR + content_url() for reliable URL-to-filesystem mapping instead of ABSPATH.
• Fix: Copy Log button now checks window.isSecureContext === true (strict boolean) before attempting the Clipboard API — previously !== false passed undefined as truthy on some environments, causing a silent failure on non-HTTPS admin URLs.

1.6.2

• Fix: Schema JSON validation was rejecting otherwise-correct JSON-LD pasted from a DOCX because Word commonly stores indentation/spacing as non-breaking spaces (U+00A0) rather than plain spaces — valid to the eye, but not valid JSON whitespace per RFC 8259, so json_decode() reported “Syntax error” even on visually-correct JSON. The extracted JSON is now normalized (NBSP and several other invisible Unicode space variants converted to plain ASCII space/newline) before validation.

1.6.1

• Fix: Find & Replace reported “Updated” for a row even when Find and Replace were identical text (a common no-op row, or one whose only real difference was a hyperlink that didn’t survive DOCX import) — str_replace()’s match count was being treated as a real-change count. The report/console now correctly show “No changes: Find and Replace produce identical text — nothing to update” instead of a false “Updated”, and the Elementor path no longer writes to the database when nothing actually changed.
• Added: DOCX import now preserves bold/italic per Word run (not just hyperlinks), and Find & Replace uses it to decide formatting intent — if a suggestion doesn’t mention formatting, any existing / wrapping the matched live text is preserved around the replacement; if the doc explicitly shows Find as bold and Replace as not (or vice versa), that explicit change is applied instead.

1.6.0

• Added: real “Download sample DOCX” buttons for Find & Replace and Schema (the two tools that accept DOCX), generated on the fly in the exact format their parsers expect — including worked examples of adding, removing, and repointing a Word hyperlink for Find & Replace, since that can’t be shown in a CSV. Schema also gained a “Download sample CSV” button it was previously missing entirely.
• Fix: Schema’s DOCX “Page URL”/”URL”/”Page” line detection had the same single-character-separator limitation as Find & Replace — hardened to accept colon, hyphen, en dash, or em dash in any combination.

1.5.9

• Fix: the DOCX “Page URL:”/”Find With:”/”Replace With:” label parser required an exact colon separator, so real-world docs broke it two different ways — “Page URL: – https://…” (colon plus a dash before the value corrupted the captured URL/text with a leading “- “) and “URL – https://…” (an en dash instead of a colon meant the label was never recognized at all, so nothing parsed). The separator is now tolerant of a colon, hyphen, en dash, or em dash in any combination, so both styles parse correctly.

1.5.8

• Improved: the “Yoast SEO scores cleared” notice after a SEO Meta run previously only said scores would update “the next time each post is opened in the editor” — requiring opening every post individually. It now links directly to Yoast’s own SEO → Tools page, where the “SEO data” section’s “Start SEO data optimization” button recalculates every post’s score server-side in one click, without opening them one by one. (Rank Math already had this same direct link; Yoast was missing it.)

1.5.7

• Fix: the “Copy log” button previously showed “Copied!” even when the copy silently failed (e.g. non-HTTPS admin, an unfocused document, or an older browser blocking the Clipboard API), since the fallback path didn’t check whether document.execCommand(“copy”) actually succeeded. It now verifies success, and if programmatic copy truly isn’t possible, highlights the console text and shows “Selected — press Ctrl+C” so you can still copy it manually instead of the button doing nothing.

1.5.6

• Fix: SEO Meta’s column auto-detection now recognizes “keyphrase” as a synonym for “keyword” (Yoast’s own UI calls this field “Focus keyphrase,” not “Focus keyword”) — files using that header were previously rejected with “no column header containing title/desc/keyword.” Also, when a file has both a “before” column (Current/Existing/Old Focus Keyphrase) and a “target” column (Recommended/New/Suggested), the target column is now always the one written — the before column is explicitly excluded, matching the same rule already used for Title/Description columns.

1.5.5

• Fix: SEO Meta, Alt Text, Link Fixer, Anchor Text Fixer, and H1 Fixer all accept .xlsx server-side (in addition to .csv) but their file picker only allowed selecting .csv and the button said “Load CSV”. The file input now allows .xlsx too, and the button reads “Load File” on every tab (matching Schema, which already accepted CSV/XLSX/DOCX).

1.5.4

• Fix: Alt Text previously reported “Not found” for any row where nothing changed, even when the Media Library attachment was found successfully but its alt text already matched the CSV’s target value (e.g. set correctly by a prior run). It now distinguishes that case — reported as “Already up to date” / skipped — from a genuine no-match, so an existing, correctly-tagged image no longer looks like a matching failure.

1.5.3

• Fix: Alt Text’s Media Library lookup (seobulkkit_alt_find_attachment_id_by_url) previously stripped “-scaled” from the search filename and required an exact tail match, so it missed images where WordPress’s own big-image auto-scaling made “-scaled” a permanent part of the primary attached file’s path (not just a resized derivative). The lookup now matches on filename stem + extension with the suffix optional either way, so these rows are found correctly instead of reporting “could not be matched anywhere.”

1.5.2

• Improved: the live console now shows what was actually written, not just that a row succeeded — SEO Meta shows which of Title/Description/Keyword changed and their new text; Alt Text shows the new alt text; Canonical shows the new canonical URL; Link Fixer/Anchor Text Fixer/Find & Replace show the old → new value; H1 Fixer shows the new heading text.

1.5.1

• Fix: URL resolution (used by SEO Meta, Find & Replace, Canonical, Schema, H1 Fixer, Link Fixer, and Anchor Text Fixer) now strips a stray trailing period/comma/semicolon/colon/exclamation/question-mark from a URL before resolving it — a common artifact when a URL was the last word of a sentence in a source document (e.g. “…page-slug/.” instead of “…page-slug/”), which previously made the row fail with “URL could not be resolved to a post.”

1.5.0

• Added: every batch tool (SEO Meta, Alt Text, Link Fixer, Anchor Text Fixer, H1 Fixer, Find & Replace, Canonical, Schema) now shows a live scrolling “Live progress” console while a batch job runs — each page/URL as it’s processed, with its outcome (updated/skipped/not-found) and a timestamp. Includes a running elapsed-time indicator and a “Copy log” button to copy the full session log to the clipboard.

1.4.2

• Improved: when a Find & Replace “Find” value contains an fragment (e.g. from a DOCX interlinking doc, for removing or repointing an existing link), it’s now matched by href + visible text using a tolerant regex instead of requiring an exact literal HTML match. Extra attributes (target, rel, class), different attribute order, and Elementor’s JSON-escaped quotes/slashes are all tolerated. Plain text-only Find/Replace rows are unaffected — same exact behavior as before.

1.4.1

• Fix: the CSV/XLSX/DOCX importer for Find & Replace previously read only plain visible text out of uploaded DOCX files, silently discarding any Word hyperlinks. For “interlinking suggestion” documents where the “Replace With” paragraph differs from “Find With” only by a hyperlink wrapped around one phrase, this meant Find and Replace came out byte-identical — the tool reported “Updated” for every row but never actually inserted a link anywhere. The DOCX parser now resolves each against the document’s relationships and embeds it as a real in the extracted text, so these interlinks are now actually inserted.

1.4.0

• Added: Find & Replace now shows a live scrolling console during Process One Batch / Auto-Run, listing each page URL as it’s processed with its outcome (updated / skipped / not matched) — previously the UI only showed an aggregate percentage with no per-page detail.

1.3.9

• Fix: seobulkkit_clear_page_caches() now also purges WP Rocket (rocket_clean_domain / rocket_clean_minify), which was missing from the supported cache-plugin list. This is the specific cache plugin running on sites where the report said “Replaced”/”Fixed” but the live page kept showing the old broken link — the database write was correct, WP Rocket’s cached HTML just was never invalidated.

1.3.8

• Fix: Link Fixer (and Anchor Text Fixer’s URL matching) now tolerates a trailing empty hash fragment (“#” or “#/”) on the live href when matching against the “Broken Link” URL from the CSV. Some themes/page builders append “#/” to links; previously that made the match exact-fail and the row was reported “not found” even though the link was really there.

1.3.7

• Fix: Link Fixer, Alt Text, Anchor Text, Find & Replace, Canonical URL, and H1 Fixer jobs now purge known page-cache plugins (WP Fastest Cache, W3 Total Cache, WP Super Cache, LiteSpeed Cache, Autoptimize) when a batch finishes, matching what the SEO Meta and Schema tools already did. Previously the database was updated correctly (and the report showed “Updated”/”Fixed”), but a caching plugin could keep serving the old cached page, making it look like nothing changed on the live site.
• Improved: redesigned the admin dashboard — icon header, pill-style tab navigation with per-tool icons, card-based panels, and polished progress bars/info boxes across all nine tools.

1.0.0

• Initial release.
• SEO Meta Importer: supports Yoast, Rank Math, All in One SEO, SEOPress, SmartCrawl, The SEO Framework.
• Image Alt Text Importer: updates Media Library alt field and inline img tags.
• Link Fixer: replace or remove links in post content, Elementor templates, and WordPress menus.
• Anchor Text Fixer: match a link by anchor text and URL, then update its anchor text and/or destination.
• Batched processing with pause/resume and downloadable CSV reports.